{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreif7s2wgyd3wf4mg35uk4m72cco7vyel6rcgopckqhloz6ek4rh4hy",
"uri": "at://did:plc:3fychdutjjusoqeq24ljch6q/app.bsky.feed.post/3mhp6spi6bmk2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreiflo6xt7is6b2iafwghkjahlgggocme5jwjsbeuqqwcywuvjhmszm"
},
"mimeType": "image/png",
"size": 24783
},
"path": "/abs/2603.19460v1",
"publishedAt": "2026-03-23T00:00:00.000Z",
"site": "https://arxiv.org",
"tags": [
"Tianyu Bell Pan",
"Damon L. Woodard"
],
"textContent": "**Authors:** Tianyu Bell Pan, Damon L. Woodard\n\nLarge language models (LLMs) demonstrate strong performance, but they often lack transparency. We introduce GeoLAN, a training framework that treats token representations as geometric trajectories and applies stickiness conditions inspired by recent developments related to the Kakeya Conjecture. We have developed two differentiable regularizers, Katz-Tao Convex Wolff (KT-CW) and Katz-Tao Attention (KT-Attn), that promote isotropy and encourage diverse attention. Our experiments with Gemma-3 (1B, 4B, 12B) and Llama-3-8B show that GeoLAN frequently maintains task accuracy while improving geometric metrics and reducing certain fairness biases. These benefits are most significant in mid-sized models. Our findings reveal scale-dependent trade-offs between geometric precision and performance, suggesting that geometry-aware training is a promising approach to enhance mechanistic interpretability.",
"title": "GeoLAN: Geometric Learning of Latent Explanatory Directions in Large Language Models"
}